C-Reactive Protein and Gamma-Glutamyltransferase
Total Page:16
File Type:pdf, Size:1020Kb
Hindawi Publishing Corporation Experimental Diabetes Research Volume 2010, Article ID 761715, 8 pages doi:10.1155/2010/761715 Clinical Study C-Reactive Protein and Gamma-Glutamyltransferase Concentrations in Relation to the Prevalence of Type 2 Diabetes Diagnosed by Glucose or HbA1c Criteria in Chinese Adults in Qingdao, China J. Ren,1 Z. C. Pang,2 W. G. Gao,3, 4, 5 H. R. Nan,6 S. J. Wang,2 L. Zhang,3, 4, 5 and Q. Qiao3, 4 1 Department of Epidemiology and Health Statistics, Shandong University, Jinan 250012, China 2 Department of Non-Communicable Disease Prevention, Qingdao Municipal Centre for Disease Control and Prevention, no. 175, Shandong Road, Qingdao 266071, China 3 Department of Public Health, University of Helsinki, 00014 Helsinki, Finland 4 Diabetes Unit, Department of Chronic Disease Prevention, National Institute for Health and Welfare, 00014 Helsinki, Finland 5 Qingdao Endocrinology and Diabetes Hospital, 266071 Qingdao, China 6 Hong Kong Institute of Diabetes and Obesity, Hong Kong, China CorrespondenceshouldbeaddressedtoZ.C.Pang,[email protected] Received 13 August 2010; Revised 12 October 2010; Accepted 12 October 2010 Academic Editor: Giuseppe Paolisso Copyright © 2010 J. Ren et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Aims. To investigate the association of C-reactive protein (CRP) and gamma glutamyltransferase (GGT) concentrations with newly diagnosed diabetes defined by either glucose or HbA1c criteria in Chinese adults. Methods. A population-based cross-sectional study was conducted in 2006. Data from 1167 men and 1607 women aged 35–74 years were analyzed. Diabetes was defined according to either glucose or HbA1c criteria alone. Results. Compared with nondiabetes, multivariate-adjusted OR (95%CI) was 1.13 (0.90,1.42) in men and 1.21 (1.00,1.45) in women for CRP and 1.42 (1.18,1.72) and 1.57 (1.31,1.87) for GGT, respectively. Neither CRP nor GGT was associated with the presence of diabetes defined by the HbA1c criterion. Conclusions. The effect of elevated CRP on diabetes defined by the glucose criterion was mediated through obesity, but elevated GGT was an independent risk factor for diabetes in this Chinese population. None of the two was, however, associated with the elevated HbA1c concentrations. 1. Introduction C-reactive protein (CRP) is a nonspecific biomarker of acute inflammation and is produced primarily in the A new large Chinese national survey [1] is in consistent with liver. Several prospective studies had shown that serum previous report from China [2] showing that type 2 diabetes CRP accelerate or increase the development of diabetes has become a seriously public health threat in China. The [5–9], particularly in women [8, 9]. There is increasing studies suggest that China is taking over India and becoming evidence showing that liver enzymes, such as gamma- the epicenter of diabetes in the world. In these previous glutamyltransferase (GGT), used as a marker of alcohol studies, type 2 diabetes has been defined according to fasting consumption or liver disease, show a dose-response relation (FPG) and/or 2-hour plasma glucose criteria after ingesting with incident diabetes even within its normal range [10]and 75 g oral glucose load (OGTT) [3]. Recently haemoglobin may also predict the development of diabetes in both genders A1c (HbA1c) test has been adopted as a diagnostic criterion independent of traditionally risk factors [11, 12]. for diabetes [4]. The impact of changes in diagnostic criteria However, type 2 diabetes in all above-mentioned studies on prevalence of diabetes and differences in phenotypes was diagnosed by FPG and/or OGTT not by HbA1c criteria. detected by different criteria are, however, still less known. In this study, the association of CRP or GGT with type 2 Experimental Diabetes Research 2 diabetes diagnosed by either glucose or HbA1c criteria performance liquid chromatography (HPLC) method [13]. is investigated based on a cross-sectional population-based The calculated HbA1c was subsequently used in the data Chinese study in Qingdao, China. analysis. The reference range for the calculated HbA1c was 4.3%–5.8%. Fasting insulin concentration was measured in 2. Subjects and Methods 2125 individuals using the chemiluminescence immunoassay method (Abbott AxSym). Fatty Liver Index (FLI)-based 2.1. Study Population. A total of 6100 individuals aged 35–74 on GGT, TG, BMI and WC-, was calculated as follows years who had lived in Qingdao City for at least 5 years were [14]: recruited in 2006 with stratified random cluster sampling FLI = (e0.953∗loge (triglycerides) + 0.139∗BMI + from 3 urban districts (Shinan, Shibei and Sifang) and 3 rural 0.718∗loge (ggt) +0.053∗waist circumference − 15.745)/(1 districts (Huangdao, Jiaonan and Jimo). Among these, 5355 + e0.953∗loge (triglycerides) + 0.139∗BMI+0.718∗loge individuals participated in the study, with a response rate (ggt) + 0.053∗waist circumference − 15.745)∗100. of 87.8%. The inclusion criteria for the current study were An FLI ≤ 30 rules out and an FLI ≥ 60 indicates hepatic (1) newly diagnosed diabetes by both FPG and/or 2 hours steatosis as detected by ultrasoundgraphy. plasma glucose criteria (2) no data missing for body mass The Ethics Committee of Qingdao Municipal Hospital index (BMI), waist circumference (WC), blood pressure approved the study. Verbal or written consent was obtained measurements, CRP, GGT, and HbA1c. A total of 2774 (1167 from each participant prior to the data collection. men) subjects with full required information were included. Height and weight were measured with participants 2.2. Classification of Diabetes. Diabetes was defined accord- wearing light clothes and without shoes, and BMI was ing to the 2006 World Health Organization (WHO)/Inter- calculated by dividing weight (Kg) by height (m) squared. national Diabetes Federation (IDF) criteria [3]andInter- WC was measured at the middle point between the rib cage national Expert Committee Report on the role of the and top of the iliac crest to the nearest 0.1 cm. Three consec- HbA1c assay for diabetes [4]. Subjects who reported a utive blood pressure readings, apart by at least 30 seconds, history of diabetes and who were under treatment with were taken from the right arm of seated subjects, and the either insulin or oral antidiabetic agents were considered average of the three readings was used in the data analysis. as previously diagnosed diabetes, regardless of their fasting Smoking status was classified as current smokers (smoking plasma glucose levels. Diagnosed diabetes was excluded daily regardless of the amount and type of smoking) and from the data analysis to reduce the potential effluence nonsmokers (including ex-smoking, smoking now and then of the anti-diabetic medications and the diabetic com- and not smoking at all). Alcohol-drinking status was defined plications. Newly diagnosed diabetes was defined if there as current drinkers (drink frequently regardless of the was an FPG level of ≥7.0 mmol/L and/or 2hPG level amount and type of alcohol) and nondrinkers (including of ≥11.1 mmol/L regardless of the HbA1c concentration. ex-drinking, drinking now and then or not drinking at Alternatively, newly diagnosed diabetes was determined all). Family history of diabetes was defined as having at if the calculated HbA1c ≥6.5% in spite of the glucose least one of parents, sibling or offspring with diabetes. levels. Waist circumference ≥90 cm for Chinese men and School years were divided into two levels (≤9or>9school ≥80 cm for Chinese women was defined as central obesity years). Occupational activities were categorized into light according to the 2004 IDF definition of metabolic syndrome (managerial staff), moderate (teacher/doctor/nurse), and [15]. heavy (worker or farmer). All subjects without a prior history of diabetes underwent a standard OGTT. Blood samples were 2.3. Statistical Analysis. A chi-square test for categorical transported in a dark box with ice to the laboratory no more variables and the general linear model (GLM) procedure than 6 hours after drawing and stored at −80◦C. The lab for continuous variables were used to compare differences assays were performed in the central laboratory of Qingdao in prevalence and in age-adjusted means among different Hiser Medical Center using Olympus AU analyzers (Olym- glucose categories. The linear association of CRP and GGT pus, Tokyo, Japan). Plasma glucose levels were determined with plasma glucose and HbA1c was tested using linear by the glucose oxidase method. Fasting serum triglycerides regression model adjusting for age, school years, family (TG) and total cholesterol (TC) were determined by enzy- history of diabetes, WC, LDL, TG, and SBP. Stepwise matic method while fasting serum CRP and high-density logistic regression analysis was performed to estimate the lipoprotein cholesterol (HDL-C) by direct method. Low- odds ratio (OR) and 95% confidence intervals (CI) for density lipoprotein cholesterol (LDL-C) was calculated using the prevalence of diabetes corresponding to a one standard the Friedewald equation. GGT and alanine aminotransferase deviation (SD) increase in continuous variables. Age, school (ALT) were measured by using an International Federation years, alcohol-drinking, smoking, family history of diabetes, of Clinical Chemistry (IFCC) method. HbA1c was measured BMI, SBP, LDL, TG, TC, CRP and GGT were fitted in the using an immunoturbidimetry method (Tina-qu.a A1C HIT multivariate model. For all analyses, variables with skewed 917 large; Roche Diagnostics). The HbA1c concentration distribution, such as CRP, and GGT, were log transformed was calculated by using the formula provided by Roche before data analysis. All analyses were performed using Diagnostics: [calculated HbA1c (%) = 0.81 × HbA1c (test SPSS (Version 15.0; SPSS Inc, Chicago, IL, USA). A P- result) + 2.39] to match the values with those found in the value less than .05 (two tailed) was considered statistically new IFCC standardization procedure, a conventional high significant.